Search

In this paper, we establish the global existence of Lagrangian solutions to the ionic Vlasov--Poisson system under mild integrability assumptions on the initial data. Our approach involves proving the well-posedness of the Poisson--Boltzmann equation for densities in $L^p$ with $p>1$, introducing a novel decomposition technique that ensures uniqueness, stability, and improved bounds for the thermalized electron density. Using this result, we construct global-in-time Lagrangian solutions while demonstrating that the energy functional remains uniformly bounded by its initial value. Additionally, we show that renormalized solutions coincide with Lagrangian solutions, highlighting the transport structure of the system, and prove that renormalized solutions coincide with weak solutions under additional integrability assumptions., Comment: 36 pages

We present a Pontryagin-Guided Direct Policy Optimization (PG-DPO) method that scales dynamic portfolio choice -- including consumption and multi-asset investment -- to tens of thousands of risky assets. By combining neural-network controls with Pontryagin's Maximum Principle (PMP), we avoid intractable dynamic programming (DP) grids, which traditionally cannot handle more than six assets in practice. Instead of approximating the value function (as in deep backward stochastic differential equation (BSDE) methods), we track a policy-fixed adjoint process and align each gradient update with continuous-time PMP conditions. A ``one-shot'' variant deploys Pontryagin controls after a brief warm-up, often achieving 100--1,000-fold accuracy improvements over naive baselines. Crucially, on modern GPUs, 100,000 iterations typically take under an hour, while a few thousand iterations (often just 1--2 minutes) already yield near-optimal solutions for portfolios with thousands of assets. Numerical experiments confirm consistency with one-asset Merton benchmarks and tractability up to 10,000 assets, surpassing the longstanding DP-based limit of fewer than seven. This enables truly large-scale continuous-time portfolio optimization.

Foundation models for computational pathology have shown great promise for specimen-level tasks and are increasingly accessible to researchers. However, specimen-level models built on these foundation models remain largely unavailable, hindering their broader utility and impact. To address this gap, we developed SpinPath, a toolkit designed to democratize specimen-level deep learning by providing a zoo of pretrained specimen-level models, a Python-based inference engine, and a JavaScript-based inference platform. We demonstrate the utility of SpinPath in metastasis detection tasks across nine foundation models. SpinPath may foster reproducibility, simplify experimentation, and accelerate the adoption of specimen-level deep learning in computational pathology research.

The expansion of large language models (LLMs) with hundreds of billions of parameters presents significant challenges to computational resources, particularly data movement and memory bandwidth. Long-context LLMs, which process sequences of tens of thousands of tokens, further increase the demand on the memory system as the complexity in attention layers and key-value cache sizes is proportional to the context length. Processing-in-Memory (PIM) maximizes memory bandwidth by moving compute to the data and can address the memory bandwidth challenges; however, PIM is not necessarily scalable to accelerate long-context LLM because of limited per-module memory capacity and the inflexibility of fixed-functional unit PIM architecture and static memory management. In this work, we propose LoL-PIM which is a multi-node PIM architecture that accelerates long context LLM through hardware-software co-design. In particular, we propose how pipeline parallelism can be exploited across a multi-PIM module while a direct PIM access (DPA) controller (or DMA for PIM) is proposed that enables dynamic PIM memory management and results in efficient PIM utilization across a diverse range of context length. We developed an MLIR-based compiler for LoL-PIM extending a commercial PIM-based compiler where the software modifications were implemented and evaluated, while the hardware changes were modeled in the simulator. Our evaluations demonstrate that LoL-PIM significantly improves throughput and reduces latency for long-context LLM inference, outperforming both multi-GPU and GPU-PIM systems (up to 8.54x and 16.0x speedup, respectively), thereby enabling more efficient deployment of LLMs in real-world applications., Comment: 15 pages, 12 figures

When trained with severely imbalanced data, deep neural networks often struggle to accurately recognize classes with only a few samples. Previous studies in long-tailed recognition have attempted to rebalance biased learning using known sample distributions, primarily addressing different classification difficulties at the class level. However, these approaches often overlook the instance difficulty variation within each class. In this paper, we propose a difficulty-aware balancing margin (DBM) loss, which considers both class imbalance and instance difficulty. DBM loss comprises two components: a class-wise margin to mitigate learning bias caused by imbalanced class frequencies, and an instance-wise margin assigned to hard positive samples based on their individual difficulty. DBM loss improves class discriminativity by assigning larger margins to more difficult samples. Our method seamlessly combines with existing approaches and consistently improves performance across various long-tailed recognition benchmarks.

Due to the subjective nature of current clinical evaluation, the need for automatic severity evaluation in dysarthric speech has emerged. DNN models outperform ML models but lack user-friendly explainability. ML models offer explainable results at a feature level, but their performance is comparatively lower. Current ML models extract various features from raw waveforms to predict severity. However, existing methods do not encompass all dysarthric features used in clinical evaluation. To address this gap, we propose a feature extraction method that minimizes information loss. We introduce an ASR transcription as a novel feature extraction source. We finetune the ASR model for dysarthric speech, then use this model to transcribe dysarthric speech and extract word segment boundary information. It enables capturing finer pronunciation and broader prosodic features. These features demonstrated an improved severity prediction performance to existing features: balanced accuracy of 83.72%., Comment: Accepted to SLT 2024

In this paper, we study the problem of generating structured objects that conform to a complex schema, with intricate dependencies between the different components (facets) of the object. The facets of the object (attributes, fields, columns, properties) can be a mix of short, structured, type-constrained facts, or long natural-language descriptions. The object has to be self-consistent between the different facets in the redundant information it carries (relative consistency), while being grounded with respect to world knowledge (absolute consistency). We frame the problem as a Language Modeling problem (Structured Object Language Modeling) and train an LLM to perform the task natively, without requiring instructions or prompt-engineering. We propose a self-supervised denoising method to train the model from an existing dataset of such objects. The input query can be the existing object itself, in which case the model acts as a regenerator, completing, correcting, normalizing the input, or any unstructured blurb to be structured. We show that the self-supervised denoising training provides a strong baseline, and that additional supervised fine-tuning with small amount of human demonstrations leads to further improvement. Experimental results show that the proposed method matches or outperforms prompt-engineered general-purpose state-of-the-art LLMs (Claude 3, Mixtral-8x7B), while being order-of-magnitude more cost-efficient.

In the world of relational database management, the query optimizer is a critical component that significantly impacts query performance. To address the challenge of optimizing query performance due to the complexity of optimizers -- especially with join operations -- we introduce Jovis. This novel visualization tool provides a window into the often intricate process of query optimization in PostgreSQL, making it more accessible and understandable. PostgreSQL employs two different query optimization strategies: the Dynamic Programming (DP) Optimizer for most scenarios and the Genetic Query Optimizer (GEQO) for more complex queries with numerous joins, both of which are supported in Jovis. Our tool visualizes the optimizer's decision-making process, from evaluating access paths for each relation to determining join orderings, all using data derived from the optimizer's logs. Jovis not only clarifies the query optimization process through visualizations but also serves as an invaluable learning tool for learners and a practical resource for experienced database professionals looking to optimize their query performance or even the query optimizer itself. The source code has been made available at https://github.com/snu-jovis.

The stellar initial mass function (IMF) is a key parameter to understand the star formation process and the integrated properties of stellar populations in remote galaxies. We present a spectroscopic study of young massive clusters (YMCs) in the starburst galaxies NGC 4038/39. The integrated spectra of seven YMCs obtained with GMOS-S attached to the 8.2-m Gemini South telescope reveal the spectral features associated with stellar ages and the underlying IMFs. We constrain the ages of the YMCs using the absorption lines and strong emission bands from Wolf-Rayet stars. The internal reddening is also estimated from the strength of the Na I D absorption lines. Based on these constraints, the observed spectra are matched with the synthetic spectra generated from a simple stellar population model. Several parameters of the clusters including age, reddening, cluster mass, and the underlying IMF are derived from the spectral matching. The ages of the YMCs range from 2.5 to 6.5 Myr, and these clusters contain stellar masses ranging from 1.6 X 10^5 M_sun to 7.9 X 10^7 M_sun. The underlying IMFs appear to differ from the universal form of the Salpeter/Kroupa IMF. Interestingly, massive clusters tend to have the bottom-heavy IMFs, although the masses of some clusters are overestimated due to the crowding effect. Based on this, our results suggest that the universal form of the IMF is not always valid when analyzing integrated light from unresolved stellar systems. However, further study with a larger sample size is required to reach a definite conclusion., Comment: 18 pages, 9 figures, accepted for publication in AJ

Human image animation aims to generate a human motion video from the inputs of a reference human image and a target motion video. Current diffusion-based image animation systems exhibit high precision in transferring human identity into targeted motion, yet they still exhibit irregular quality in their outputs. Their optimal precision is achieved only when the physical compositions (i.e., scale and rotation) of the human shapes in the reference image and target pose frame are aligned. In the absence of such alignment, there is a noticeable decline in fidelity and consistency. Especially, in real-world environments, this compositional misalignment commonly occurs, posing significant challenges to the practical usage of current systems. To this end, we propose Test-time Procrustes Calibration (TPC), which enhances the robustness of diffusion-based image animation systems by maintaining optimal performance even when faced with compositional misalignment, effectively addressing real-world scenarios. The TPC provides a calibrated reference image for the diffusion model, enhancing its capability to understand the correspondence between human shapes in the reference and target images. Our method is simple and can be applied to any diffusion-based image animation system in a model-agnostic manner, improving the effectiveness at test time without additional training., Comment: 24 pages, 16 figures, NeurIPS 2024

Verifying the violation of Bell's inequality is one of the most representative methods to demonstrate that entangled photon pairs prepared in a quantum optics-based system exhibit quantum properties. While experiments on Bell inequality violations have been theoretically well-established and extensively conducted to implement various quantum information technologies in laboratory settings, mathematical modeling for accurately predicting the distribution of high-intensity entangled photon pairs in high-loss environments remains an issue that requires further research. As the brightness of the entangled photon pairs increases, the influence of multi-photon effects becomes more significant, leading to a decrease in the CHSH value $S$ and also a reduction in the standard deviation of the CHSH value $\Delta S$. Therefore, a new analysis of the $(S-2)/\Delta S$ value is required to more precisely confirm the degree of CHSH inequality violation including the reliability of $S$. In this paper, we propose a mathematical model to predict the $(S-2)/\Delta S$ value as a function of the brightness of the entangled photon pair source, and we also suggest the need to optimize the brightness of this source. Additionally, we provide experimental evidence supporting this model. The experiment confirms that when the mean photon number is $\mu=0.026$ in an entanglement distribution setup with a total loss of $-19.03$ dB, the CHSH value drops to 2.69, while the $(S-2)/\Delta S$ value increases to 60.95., Comment: 11 pages, 1 table, and 4 figures

Federated fine-tuning for Large Language Models (LLMs) has recently gained attention due to the heavy communication overhead of transmitting large model updates. Low Rank Adaptation (LoRA) has been proposed as a solution, yet its application in federated learning is complicated by discordance in aggregation. Existing methods addressing this discordance often suffer from performance degradation at low ranks in heterogeneous data settings. In response, we introduce LoRA-A2 (Low Rank Adaptation with Alternating freeze and Adaptive rank selection), which demonstrates robustness in challenging settings with low ranks and high data heterogeneity. Our experimental findings reveal that LoRA-A2 maintains performance even under extreme heterogeneity and low rank conditions, achieving up to a 99.8% reduction in uploaded parameters compared to full fine-tuning without compromising performance. This adaptive mechanism boosts robustness and communication efficiency in federated fine-tuning, enabling the practical deployment of LLMs in resource-constrained environments.

Keyphrase ranking plays a crucial role in information retrieval and summarization by indexing and retrieving relevant information efficiently. Advances in natural language processing, especially large language models (LLMs), have improved keyphrase extraction and ranking. However, traditional methods often overlook diversity, resulting in redundant keyphrases. We propose a novel approach using Submodular Function Optimization (SFO) to balance relevance and diversity in keyphrase ranking. By framing the task as submodular maximization, our method selects diverse and representative keyphrases. Experiments on benchmark datasets show that our approach outperforms existing methods in both relevance and diversity metrics, achieving SOTA performance in execution time. Our code is available online.

Despite the success of Large Language Models (LLMs), they still face challenges related to high inference costs and memory requirements. To address these issues, Knowledge Distillation (KD) has emerged as a popular method for model compression, with student-generated outputs (SGOs) being particularly notable for reducing the mismatch between training and inference. However, SGOs often produce noisy and biased sequences, which can lead to misguidance from the teacher model, especially in long sequences. To mitigate these challenges, we propose SWITCH (Studying WIth TeaCHer for Knowledge Distillation), a novel approach that strategically incorporates the teacher model during the student's sequence generation. SWITCH identifies discrepancies between the token probabilities of the teacher and student models, allowing the teacher to intervene selectively, particularly in long sequences that are more prone to teacher misguidance. Extensive experimental results across three model families and five instruction-following datasets show that SWITCH surpasses traditional KD methods, particularly excelling in the generation of long sequential data.

Deep State Space Models (SSMs), such as Mamba (Gu & Dao, 2024), have emerged as powerful tools for language modeling, offering high performance with efficient inference and linear scaling in sequence length. However, the application of parameter-efficient fine-tuning (PEFT) methods to SSM-based models remains largely unexplored. This paper aims to systematically study two key questions: (i) How do existing PEFT methods perform on SSM-based models? (ii) Which modules are most effective for fine-tuning? We conduct an empirical benchmark of four basic PEFT methods on SSM-based models. Our findings reveal that prompt-based methods (e.g., prefix-tuning) are no longer effective, an empirical result further supported by theoretical analysis. In contrast, LoRA remains effective for SSM-based models. We further investigate the optimal application of LoRA within these models, demonstrating both theoretically and experimentally that applying LoRA to linear projection matrices without modifying SSM modules yields the best results, as LoRA is not effective at tuning SSM modules. To further improve performance, we introduce LoRA with Selective Dimension tuning (SDLoRA), which selectively updates certain channels and states on SSM modules while applying LoRA to linear projection matrices. Extensive experimental results show that this approach outperforms standard LoRA., Comment: Code is available at https://github.com/furiosa-ai/ssm-peft

Recent state-of-the-art neural audio compression models have progressively adopted residual vector quantization (RVQ). Despite this success, these models employ a fixed number of codebooks per frame, which can be suboptimal in terms of rate-distortion tradeoff, particularly in scenarios with simple input audio, such as silence. To address this limitation, we propose variable bitrate RVQ (VRVQ) for audio codecs, which allows for more efficient coding by adapting the number of codebooks used per frame. Furthermore, we propose a gradient estimation method for the non-differentiable masking operation that transforms from the importance map to the binary importance mask, improving model training via a straight-through estimator. We demonstrate that the proposed training framework achieves superior results compared to the baseline method and shows further improvement when applied to the current state-of-the-art codec., Comment: Accepted at NeurIPS 2024 Workshop on Machine Learning and Compression

JWST observations of the young Galactic supernova remnant Cassiopeia A revealed an unexpected structure seen as a green emission feature in colored composite MIRI F1130W and F1280W images - hence dubbed the Green Monster - that stretches across the central parts of the remnant in projection. Combining the kinematic information from NIRSpec and MIRI MRS with the multi-wavelength imaging from NIRCam and MIRI, we associate the Green Monster with circumstellar material that was lost during an asymmetric mass-loss phase. MIRI images are dominated by dust emission but its spectra show emission lines from Ne, H and Fe with low radial velocities indicative of a CSM nature. An X-ray analysis of this feature in a companion paper (Vink et al. 2024) supports its CSM nature and detects significant blue shifting, thereby placing the Green Monster on the near side, in front of the Cas A SN remnant. The most striking features of the Green Monster are dozens of almost perfectly circular 1" - 3" sized holes, most likely created by interaction between high-velocity SN ejecta material and the CSM. Further investigation is needed to understand whether these holes were formed by small 8000-10500 km/s N-rich ejecta knots that penetrated and advanced out ahead of the remnant's 5000 - 6000 km/s outer blastwave, or by narrow ejecta fingers that protrude into the forward-shocked CSM. The detection of the Green Monster provides further evidence of the highly asymmetric mass-loss that Cas A's progenitor star underwent prior to explosion., Comment: 28 pages, 12 figures, resubmitted to ApJL after minor revision, comments welcome

Open Radio Access Networks (RAN) aim to bring a paradigm shift to telecommunications industry, by enabling an open, intelligent, virtualized, and multi-vendor interoperable RAN ecosystem. At the center of this movement, O-RAN ALLIANCE defines the O-RAN architecture and standards, so that companies around the globe can use these specifications to create innovative and interoperable solutions. To accelerate the adoption of O-RAN products, rigorous testing of O-RAN Radio Unit (O-RU) and other O-RAN products plays a key role. O-RAN ALLIANCE has approved around 20 Open Testing and Integration Centres (OTICs) globally. OTICs serve as vendor-neutral platforms for providing the testing and integration services, with the vision that an O-RAN product certified in any OTIC is accepted in other parts of the world. To demonstrate the viability of such a certified-once-and-use-everywhere approach, one theme in the O-RAN Global PlugFest Spring 2024 is to demonstrate consistent and repeatable testing for the open fronthaul interface across multiple labs. Towards this, Japan OTIC and Asia Pacific OTIC in Singapore have teamed up together with an O-RU vendor and Keysight Technology. Our international team successfully completed all test cases defined by O-RAN ALLIANCE for O-RU conformance testing. In this paper, we share our journey in achieving this outcome, focusing on the challenges we have overcome and the lessons we have learned through this process., Comment: Published version at RitiRAN Workshop - co-located with IEEE VTC Fall 2024

Materials hosting flat bands at the vicinity of the Fermi level promote exotic symmetry broken states. Common to many of these are van Hove singularities at saddle points of the dispersion or even higher-order van Hove singularities where the dispersion is flattened further. The band structure of kagome metals hosts both a flat band and two regular saddle points flanking a Dirac node. We investigate the kagome ferromagnetic metal Co$_3$Sn$_2$S$_2$ using scanning tunneling spectroscopy. We identify a new mechanism by which a triangular distortion on its kagome Co$_3$Sn surface termination considerably flattens the saddle point dispersion, and induces an isolated higher-order van Hove singularity (HOvHS) with algebraically divergent density of states pinned to the Fermi energy. The distortion-induced HOvHS precipitates a Pomeranchuk instability of the Fermi surface, resulting in the formation of a series of nematic electronic states. We visualize the nematic order across an energy shell of about 100 meV in both real-, reciprocal-, and momentum-spaces, as a cascade of wavefunction distributions which spontaneously break the remaining rotational symmetry of the underlying distorted kagome lattice, without generating any additional translational symmetry breaking. It signifies the spontaneous removal of a subset of saddle points from the Fermi energy to lower energies. By tracking the electronic wavefunction structure across the deformed Fermi surface we further identify a charge pumping-like evolution of the wavefunction center of mass. The mechanism we find for the generation of higher-order saddle points under a kagome distortion may be common to other kagome materials, and potentially other lattice structures, suggesting a generic new avenue for inducing unconventional electronic instabilities towards exotic states of matter.

To search for a conceptual picture describing the magnetization plateau phenomenon, we surveyed the crystal structures and the spin lattices of those magnets exhibiting plateaus in their magnetization vs. magnetic field curves by probing the three questions: (a) why only certain magnets exhibit magnetization plateaus, (b) why there occur several different types of magnetization plateaus, and (c) what controls the widths of magnetization plateaus. We show that the answers to these questions lie in how the magnets under field absorb Zeeman energy hence changing their magnetic structures. The magnetic structure of a magnet insulator is commonly described in terms of its spin lattice, which requires the determination of the spin exchanges nonnegligible strengths between the magnetic ions. Our work strongly suggests that a magnet under magnetic field partitions its spin lattice into antiferromagnetic (AFM) or ferrimagnetic fragments by breaking its weak magnetic bonds. Our supposition of the field-induced partitioning of spin lattice into magnetic fragments is supported by the anisotropic magnetization plateaus of Ising magnets and by the highly anisotropic width of the 1/3-magnetization plateau in azurite. The answers to the three questions (a) - (c) emerge naturally by analyzing how these fragments are formed under magnetic field., Comment: A survey of the magnetization plateau phenomenon; 128 pages including the Supporting Information

Text-based diffusion video editing systems have been successful in performing edits with high fidelity and textual alignment. However, this success is limited to rigid-type editing such as style transfer and object overlay, while preserving the original structure of the input video. This limitation stems from an initial latent noise employed in diffusion video editing systems. The diffusion video editing systems prepare initial latent noise to edit by gradually infusing Gaussian noise onto the input video. However, we observed that the visual structure of the input video still persists within this initial latent noise, thereby restricting non-rigid editing such as motion change necessitating structural modifications. To this end, this paper proposes Dilutional Noise Initialization (DNI) framework which enables editing systems to perform precise and dynamic modification including non-rigid editing. DNI introduces a concept of `noise dilution' which adds further noise to the latent noise in the region to be edited to soften the structural rigidity imposed by input video, resulting in more effective edits closer to the target prompt. Extensive experiments demonstrate the effectiveness of the DNI framework., Comment: 17 pages, 11 figures, ECCV 2024

SiC structures, including nanowires and films, can be effectively grown on Si substrates through carbonization. However, growth parameters other than temperature, which influence the preferential formation of SiC nanowires or films, have not yet been identified. In this work, we investigate SiC synthesis via Si carbonization using methane (CH$_4$) by varying the growth temperature and the hydrogen to methane gas flow ratio (H$_2$/CH$_4$). We demonstrate that adjusting these parameters allows for the preferential growth of SiC nanowires or films. Specifically, SiC nanowires are preferentially grown when the H$_2$/CH$_4$ ratio exceeds a specific threshold, which varies with the growth temperature, ranging between 1200$^\circ$C and 1310$^\circ$C. Establishing this precise growth window for SiC nanowires in terms of the H$_2$/CH$_4$ ratio and growth temperature provides new insights into the parameter-driven morphology of SiC. Furthermore, we propose a mechanistic model to explain the preferential growth of either SiC nanowires or films, based on the kinetics of gas-phase reactions and surface processes. These findings not only advance our understanding of SiC growth mechanisms but also pave the way for optimized fabrication strategies for SiC-based nanostructures., Comment: 6 pages, 5 figures

Music timbre transfer is a challenging task that involves modifying the timbral characteristics of an audio signal while preserving its melodic structure. In this paper, we propose a novel method based on dual diffusion bridges, trained using the CocoChorales Dataset, which consists of unpaired monophonic single-instrument audio data. Each diffusion model is trained on a specific instrument with a Gaussian prior. During inference, a model is designated as the source model to map the input audio to its corresponding Gaussian prior, and another model is designated as the target model to reconstruct the target audio from this Gaussian prior, thereby facilitating timbre transfer. We compare our approach against existing unsupervised timbre transfer models such as VAEGAN and Gaussian Flow Bridges (GFB). Experimental results demonstrate that our method achieves both better Fr\'echet Audio Distance (FAD) and melody preservation, as reflected by lower pitch distances (DPD) compared to VAEGAN and GFB. Additionally, we discover that the noise level from the Gaussian prior, $\sigma$, can be adjusted to control the degree of melody preservation and amount of timbre transferred.

We establish global existence of mild solutions to the BGK model proposed by Bouchut [J. Stat. Phys., 95, (1999), 113--170] under the minimal assumption of finite kinetic entropy initial data. Moreover we rigorously derive a kinetic entropy inequality, which combined with the theory developed by Berthelin and Vasseur [SIAM J. Math. Anal., 36, (2005), 1807--1835] leads to the hydrodynamic limit to the barotropic Euler equations. The main tools employed in the analysis are stability estimates for the Maxwellian and a velocity averaging lemma., Comment: 24 pages, 1 figure

To overcome the burden on the memory size and bandwidth due to ever-increasing size of large language models (LLMs), aggressive weight quantization has been recently studied, while lacking research on quantizing activations. In this paper, we present a hardware-software co-design method that results in an energy-efficient LLM accelerator, named OPAL, for generation tasks. First of all, a novel activation quantization method that leverages the microscaling data format while preserving several outliers per sub-tensor block (e.g., four out of 128 elements) is proposed. Second, on top of preserving outliers, mixed precision is utilized that sets 5-bit for inputs to sensitive layers in the decoder block of an LLM, while keeping inputs to less sensitive layers to 3-bit. Finally, we present the OPAL hardware architecture that consists of FP units for handling outliers and vectorized INT multipliers for dominant non-outlier related operations. In addition, OPAL uses log2-based approximation on softmax operations that only requires shift and subtraction to maximize power efficiency. As a result, we are able to improve the energy efficiency by 1.6~2.2x, and reduce the area by 2.4~3.1x with negligible accuracy loss, i.e., <1 perplexity increase., Comment: 7 pages, 8 figures, DAC2024 accepted

When discussing difficult topics online, is it common to meaningfully engage with people from diverse perspectives? Why or why not? Could features of the online environment be redesigned to encourage civil conversation across difference? In this paper, we study discussions of gun policy on Reddit, with the overarching goal of developing insights into the potential of the internet to support understanding across difference. We use two methods: a clustering analysis of Reddit posts to contribute insights about what people discuss, and an interview study of twenty Reddit users to help us understand why certain kinds of conversation take place and others don't. We find that the discussion of gun politics falls into three groups: conservative pro-gun, liberal pro-gun, and liberal anti-gun. Each type of group has its own characteristic topics. While our subjects state that they would be willing to engage with others across the ideological divide, in practice they rarely do. Subjects are siloed into like-minded subreddits through a two-pronged effect, where they are simultaneously pushed away from opposing-view communities while actively seeking belonging in like-minded ones. Another contributing factor is Reddit's "karma" mechanism: fear of being downvoted and losing karma points and social approval of peers causes our subjects to hesitate to say anything in conflict with group norms. The pseudonymous nature of discussion on Reddit plays a complex role, with some subjects finding it freeing and others fearing reprisal from others not bound by face-to-face norms of politeness. Our subjects believe that content moderation can help ameliorate these issues; however, our findings suggest that moderators need different tools to do so effectively. We conclude by suggesting platform design changes that might increase discussion across difference., Comment: CSCW 2024

Introduction: Deep learning models hold great promise for digital pathology, but their opaque decision-making processes undermine trust and hinder clinical adoption. Explainable AI methods are essential to enhance model transparency and reliability. Methods: We developed HIPPO, an explainable AI framework that systematically modifies tissue regions in whole slide images to generate image counterfactuals, enabling quantitative hypothesis testing, bias detection, and model evaluation beyond traditional performance metrics. HIPPO was applied to a variety of clinically important tasks, including breast metastasis detection in axillary lymph nodes, prognostication in breast cancer and melanoma, and IDH mutation classification in gliomas. In computational experiments, HIPPO was compared against traditional metrics and attention-based approaches to assess its ability to identify key tissue elements driving model predictions. Results: In metastasis detection, HIPPO uncovered critical model limitations that were undetectable by standard performance metrics or attention-based methods. For prognostic prediction, HIPPO outperformed attention by providing more nuanced insights into tissue elements influencing outcomes. In a proof-of-concept study, HIPPO facilitated hypothesis generation for identifying melanoma patients who may benefit from immunotherapy. In IDH mutation classification, HIPPO more robustly identified the pathology regions responsible for false negatives compared to attention, suggesting its potential to outperform attention in explaining model decisions. Conclusions: HIPPO expands the explainable AI toolkit for computational pathology by enabling deeper insights into model behavior. This framework supports the trustworthy development, deployment, and regulation of weakly-supervised models in clinical and research settings, promoting their broader adoption in digital pathology.

Keyphrase Prediction (KP) is essential for identifying keyphrases in a document that can summarize its content. However, recent Natural Language Processing (NLP) advances have developed more efficient KP models using deep learning techniques. The limitation of a comprehensive exploration jointly both keyphrase extraction and generation using pre-trained language models spotlights a critical gap in the literature, compelling our survey paper to bridge this deficiency and offer a unified and in-depth analysis to address limitations in previous surveys. This paper extensively examines the topic of pre-trained language models for keyphrase prediction (PLM-KP), which are trained on large text corpora via different learning (supervisor, unsupervised, semi-supervised, and self-supervised) techniques, to provide respective insights into these two types of tasks in NLP, precisely, Keyphrase Extraction (KPE) and Keyphrase Generation (KPG). We introduce appropriate taxonomies for PLM-KPE and KPG to highlight these two main tasks of NLP. Moreover, we point out some promising future directions for predicting keyphrases.

A high-brightness entangled photon pair (HBEPP) source is essential for conducting entanglement-based quantum key distribution (QKD) between a satellite and a ground station. While an ultrabright source can overcome significant losses in satellite-based QKD (SQKD) and increase the sifted key rate, it also induces the multi-photon effect, raising the system's error rate. To accurately estimate system performance, we first present an analytical model for calculating the measurement probabilities of HBEPP distribution through an asymmetric loss channel. Based on this model, we propose the use of a passive-intensity HBEPP source for SQKD systems, assuming a polarization-independent channel and threshold detectors for measurement. We confirm that fixing the mean photon number at $\bar{\mu}=0.1$ achieves a performance of $99.7\%$ compared to the ideal one-way communication entanglement-based SQKD protocol, which is effectively optimizing the HBEPP source brightness in accordance with system losses., Comment: 9 pages, 5 figures, 3 tables

The evolution of the Fermi surface across the quantum critical point (QCP), which is relevant for characterizing the quantum criticality and understanding its relation with unconventional superconductivity, is an intriguing subject in the study of strongly correlated electron systems. In this study, we report the thermopower measurements to investigate a change in Fermi surface across the QCP in pure and 4.4% Sn-doped CeRhIn$_5$. Results show that their thermopower behavior differs significantly in the vicinity of their respective pressure-induced QCP. In pure CeRhIn$_5$, a drastic collapse of the thermopower takes place at the Kondo breakdown QCP, where the Fermi surface reconstructs concurrently with the development of the magnetic order. By contrast, the thermopower exhibits a broadly symmetric behavior around the QCP in 4.4% Sn-doped CeRhIn$_5$, which is a characteristic of the spin-density-wave QCP. These observations are consistent with the theoretical expectations and suggest the effectiveness of thermopower measurement in discriminating the nature of quantum criticality in heavy-fermion systems., Comment: 12 pages, 4 figures

We investigate how the stellar rotational support changes as a function of spatially resolved stellar population age ($\rm D_n4000$) and relative central stellar surface density ($\Delta \Sigma_1$) for MaNGA isolated/central disk galaxies. We find that the galaxy rotational support $\lambda_{R_\mathrm{e}}$ varies smoothly as a function of $\Delta \Sigma_1$ and $\rm D_n4000$. $\rm D_n4000$ vs. $\Delta \Sigma_1$ follows a "J-shape", with $\lambda_{R_\mathrm{e}}$ contributing to the scatters. In this "J-shaped" pattern rotational support increases with central $\rm D_n4000$ when $\Delta \Sigma_1$ is low but decreases with $\Delta \Sigma_1$ when $\Delta \Sigma_1$ is high. Restricting attention to low-$\Delta \Sigma_1$ (i.e, large-radius) galaxies, we suggest that the trend of increasing rotational support with $\rm D_n4000$ for these objects is produced by a mix of two different processes, a primary trend characterized by growth in $\lambda_{R_\mathrm{e}}$ along with mass through gas accretion, on top of which disturbance episodes are overlaid, which reduce rotational support and trigger increased star formation. An additional finding is that star forming galaxies with low $\Delta \Sigma_1$ have relatively larger radii than galaxies with higher $\Delta \Sigma_1$ at fixed stellar mass. Assuming that these relative radii rankings are preserved while galaxies are star forming then implies clear evolutionary paths in central $\rm D_n4000$ vs. $\Delta \Sigma_1$. The paper closes with comments on the implications that these paths have for the evolution of pseudo-bulges vs. classical-bulges. The utility of using $\rm D_n4000$-$\Delta \Sigma_1$ to study $\lambda_{R_\mathrm{e}}$ reinforces the notion that galaxy kinematics correlate both with structure and with stellar-population state, and indicates the importance of a multi-dimensional description for understanding bulge and galaxy evolution., Comment: 24 pages, 22 figures (including Appendix), accepted for publication in MNRAS

The SPECtrophotometer for TRansmission spectroscopy of exoplanets (SPECTR) is a new low-resolution optical (3800 {\AA} - 6850 {\AA}) spectrophotometer installed at the Bohyunsan Optical Astronomy Observatory (BOAO) 1.8 m telescope. SPECTR is designed for observing the transmission spectra of transiting exoplanets. Unique features of SPECTR are its long slit length of 10 arcminutes which facilitates observing the target and the comparison star simultaneously, and its wide slit width to minimize slit losses. SPECTR will be used to survey exoplanets, such as those identified by the Transiting Exoplanet Survey Satellite (TESS), providing information about their radii across the wavelength range. In this paper, we present the design of SPECTR and the observational results of the partial transit of HD 189733 b and a full transit of Qatar-8 b. Analyses show the SPECTR's capability on the white light curves with an accuracy of one ppt. The transmission spectrum of HD 189733 b shows general agreement with previous studies., Comment: 12 pages, 11 figures, 5 tables, accepted for PASP

In this paper, we present NECHO v2, a novel framework designed to enhance the predictive accuracy of multimodal sequential patient diagnoses under uncertain missing visit sequences, a common challenge in real clinical settings. Firstly, we modify NECHO, designed in a diagnosis code-centric fashion, to handle uncertain modality representation dominance under the imperfect data. Secondly, we develop a systematic knowledge distillation by employing the modified NECHO as both teacher and student. It encompasses a modality-wise contrastive and hierarchical distillation, transformer representation random distillation, along with other distillations to align representations between teacher and student tightly and effectively. We also propose curriculum learning guided random data erasing within sequences during both training and distillation of the teacher to lightly simulate scenario with missing visit information, thereby fostering effective knowledge transfer. As a result, NECHO v2 verifies itself by showing robust superiority in multimodal sequential diagnosis prediction under both balanced and imbalanced incomplete settings on multimodal healthcare data., Comment: Accepted by ICASSP 2025 (2025 IEEE International Conference on Acoustics, Speech, and Signal Processing)

Current image editing methods primarily utilize DDIM Inversion, employing a two-branch diffusion approach to preserve the attributes and layout of the original image. However, these methods encounter challenges with non-rigid edits, which involve altering the image's layout or structure. Our comprehensive analysis reveals that the high-frequency components of DDIM latent, crucial for retaining the original image's key features and layout, significantly contribute to these limitations. Addressing this, we introduce FlexiEdit, which enhances fidelity to input text prompts by refining DDIM latent, by reducing high-frequency components in targeted editing areas. FlexiEdit comprises two key components: (1) Latent Refinement, which modifies DDIM latent to better accommodate layout adjustments, and (2) Edit Fidelity Enhancement via Re-inversion, aimed at ensuring the edits more accurately reflect the input text prompts. Our approach represents notable progress in image editing, particularly in performing complex non-rigid edits, showcasing its enhanced capability through comparative experiments., Comment: ECCV 2024

Open-source LLMs have shown great potential as fine-tuned chatbots, and demonstrate robust abilities in reasoning and surpass many existing benchmarks. Retrieval-Augmented Generation (RAG) is a technique for improving the performance of LLMs on tasks that the models weren't explicitly trained on, by leveraging external knowledge databases. Numerous studies have demonstrated the effectiveness of RAG to more successfully accomplish downstream tasks when using vector datasets that consist of relevant background information. It has been implicitly assumed by those in the field that if adversarial background information is utilized in this context, that the success of using a RAG-based approach would be nonexistent or even negatively impact the results. To address this assumption, we tested several open-source LLMs on the ability of RAG to improve their success in answering multiple-choice questions (MCQ) in the medical subspecialty field of Nephrology. Unlike previous studies, we examined the effect of RAG in utilizing both relevant and adversarial background databases. We set up several open-source LLMs, including Llama 3, Phi-3, Mixtral 8x7b, Zephyr$\beta$, and Gemma 7B Instruct, in a zero-shot RAG pipeline. As adversarial sources of information, text from the Bible and a Random Words generated database were used for comparison. Our data show that most of the open-source LLMs improve their multiple-choice test-taking success as expected when incorporating relevant information vector databases. Surprisingly however, adversarial Bible text significantly improved the success of many LLMs and even random word text improved test taking ability of some of the models. In summary, our results demonstrate for the first time the countertintuitive ability of adversarial information datasets to improve the RAG-based LLM success., Comment: 24 pages, 3 figures, 11 tables

Large Language Models (LLMs) excel in various language tasks but they often generate incorrect information, a phenomenon known as "hallucinations". Retrieval-Augmented Generation (RAG) aims to mitigate this by using document retrieval for accurate responses. However, RAG still faces hallucinations due to vague queries. This study aims to improve RAG by optimizing query generation with a query-document alignment score, refining queries using LLMs for better precision and efficiency of document retrieval. Experiments have shown that our approach improves document retrieval, resulting in an average accuracy gain of 1.6%.

In this analysis we apply a model-independent framework to test the flat $\Lambda$CDM cosmology using simulated SNIa data from the upcoming Legacy Survey of Space and Time (LSST) and combined with simulated Dark Energy Spectroscopic Instrument (DESI) five-years Baryon Acoustic Oscillations (BAO) data. We adopt an iterative smoothing technique to reconstruct the expansion history from SNIa data, which, when combined with BAO measurements, facilitates a comprehensive test of the Universe's curvature and the nature of dark energy. The analysis is conducted under three different mock true cosmologies: a flat $\Lambda$CDM universe, a universe with a notable curvature ($\Omega_{k,0} = 0.1$), and one with dynamically evolving dark energy. Each cosmology demonstrates different kinds and varying degrees of deviation from the standard model predictions. We forecast that our reconstruction technique can constrain cosmological parameters, such as the curvature ($\Omega_{k,0}$) and $c/H_0 r_\mathrm{d}$, with a precision of approximately 0.5\% for $c/H_0r_\mathrm{d}$ and 0.04 for $\Omega_{k,0}$, competitive with current cosmic microwave background constraints, without assuming any form of dark energy.

Language models (LMs) are often expected to generate strings in some formal language; for example, structured data, API calls, or code snippets. Although LMs can be tuned to improve their adherence to formal syntax, this does not guarantee conformance, especially with smaller LMs suitable for large-scale deployment. In addition, tuning requires significant resources, making it impractical for uncommon or task-specific formats. To prevent downstream parsing errors we would ideally constrain the LM to only produce valid output, but this is severely complicated by tokenization, which is typically both ambiguous and misaligned with the formal grammar. We solve these issues through the application of automata theory, deriving an efficient closed-form solution for the regular languages, a broad class of formal languages with many practical applications, including API calls or schema-guided JSON and YAML. We also discuss pragmatic extensions for coping with the issue of high branching factor, and extend our techniques to deterministic context-free languages, which similarly admit an efficient closed-form solution. Previous work on this topic (Willard and Louf, 2023) layers bespoke solutions onto automata, leading to problems with speed, correctness, and extensibility. Instead, we reformulate the entire task in terms of automata so we can leverage well-studied and well-optimized algorithms. Our system compiles constraints ~7,000x faster, is provably correct, and can be extended in a modular fashion., Comment: COLM 2024 Camera-ready version, responding to feedback from reviewers

Model Predictive Control (MPC) is an optimal control strategy suited for flood control of water resources infrastructure. Despite many studies on reservoir flood control and their theoretical contribution, optimisation methodologies have not been widely applied in real-time operation due to disparities between research assumptions and practical requirements. First, tacit objectives such as minimising the magnitude and frequency of changes in the existing outflow schedule are considered important in practice, but these are nonlinear and challenging to formulate to suit all conditions. Incorporating these objectives transforms the problem into a multi-objective nonlinear optimisation problem that is difficult to solve online. Second, it is reasonable to assume that the weights and parameters are not stationary because the preference varies depending on the state of the system. To overcome these limitations, we propose a framework that converts the original intractable problem into parameterized linear MPC problems with dynamic optimisation of weights and parameters. This is done by introducing a model-based learning concept under the assumption of the dynamic nature of the operator's preference. We refer to this framework as Parameterised Dynamic MPC (PD-MPC). The effectiveness of this framework is demonstrated through a numerical experiment for the Daecheong multipurpose reservoir in South Korea. We find that PD-MPC outperforms `standard' MPC-based designs without a dynamic optimisation process under the same uncertain inflows.

The Ice-sheet and Sea-level System Model (ISSM) provides solutions for Stokes equations relevant to ice sheet dynamics by employing finite element and fine mesh adaption. However, since its finite element method is compatible only with Central Processing Units (CPU), the ISSM has limits on further economizing computational time. Thus, by taking advantage of Graphics Processing Units (GPUs), we design a graph convolutional network (GCN) as a fast emulator for ISSM. The GCN is trained and tested using the 20-year transient ISSM simulations in the Pine Island Glacier (PIG). The GCN reproduces ice thickness and velocity with a correlation coefficient greater than 0.998, outperforming the traditional convolutional neural network (CNN). Additionally, GCN shows 34 times faster computational speed than the CPU-based ISSM modeling. The GPU-based GCN emulator allows us to predict how the PIG will change in the future under different melting rate scenarios with high fidelity and much faster computational time., Comment: 5 pages, 4 figures, submitted to the 2024 IEEE International Geoscience and Remote Sensing Symposium. arXiv admin note: text overlap with arXiv:2402.05291

Although numerical models provide accurate solutions for ice sheet dynamics based on physics laws, they accompany intensified computational demands to solve partial differential equations. In recent years, convolutional neural networks (CNNs) have been widely used as statistical emulators for those numerical models. However, since CNNs operate on regular grids, they cannot represent the refined meshes and computational efficiency of finite-element numerical models. Therefore, instead of CNNs, this study adopts an equivariant graph convolutional network (EGCN) as an emulator for the ice sheet dynamics modeling. EGCN reproduces ice thickness and velocity changes in the Helheim Glacier, Greenland, and Pine Island Glacier, Antarctica, with 260 times and 44 times faster computation time, respectively. Compared to the traditional CNN and graph convolutional network, EGCN shows outstanding accuracy in thickness prediction near fast ice streams by preserving the equivariance to the translation and rotation of graphs., Comment: 6 pages, 2 figures, submitted to the ICML 2024 Workshop on Machine Learning for Earth System Modeling

We propose a novel method for learning representations of poses for 3D deformable objects, which specializes in 1) disentangling pose information from the object's identity, 2) facilitating the learning of pose variations, and 3) transferring pose information to other object identities. Based on these properties, our method enables the generation of 3D deformable objects with diversity in both identities and poses, using variations of a single object. It does not require explicit shape parameterization such as skeletons or joints, point-level or shape-level correspondence supervision, or variations of the target object for pose transfer. To achieve pose disentanglement, compactness for generative models, and transferability, we first design the pose extractor to represent the pose as a keypoint-based hybrid representation and the pose applier to learn an implicit deformation field. To better distill pose information from the object's geometry, we propose the implicit pose applier to output an intrinsic mesh property, the face Jacobian. Once the extracted pose information is transferred to the target object, the pose applier is fine-tuned in a self-supervised manner to better describe the target object's shapes with pose variations. The extracted poses are also used to train a cascaded diffusion model to enable the generation of novel poses. Our experiments with the DeformThings4D and Human datasets demonstrate state-of-the-art performance in pose transfer and the ability to generate diverse deformed shapes with various objects and poses., Comment: NeurIPS 2024